© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 64065
    Prev
    Next
    INDEX
    Explanations

    `all` or `solely` or `mostly` followed by a descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    éĿĻéĿĻ
    -0.08
    åı£æ°Ķ
    -0.08
    endar
    -0.08
    æŃ¹
    -0.08
    éķĵ
    -0.08
    orian
    -0.08
    chars
    -0.08
    ãģ¡ãĤī
    -0.08
    smarty
    -0.08
     usu
    -0.08
    POSITIVE LOGITS
    é¾Ļ头
    0.10
     LLP
    0.09
    èIJ½åĪ°
    0.09
     ÑĢазнÑĭÑħ
    0.09
    管çIJĨä¸Ńå¿ĥ
    0.09
    纯
    0.09
    åĪ°çİ°åľ¨
    0.09
    å¼Ĥåľ°
    0.09
    ä¸į平衡
    0.09
    éĿŀçī©è´¨
    0.09
    Activations Density 0.690%

    No Known Activations